Conventionally, there has been known a technique in which (i) an image is obtained by reading a document by a scanner, (ii) a document image, which is a part of the image in which part the document is shown, is specified, and then (ii) only the document image is cropped from the image. Patent Literature 1 discloses an image input apparatus which crops a document image.
The image input apparatus disclosed in Patent Literature 1 carries out processing as follows. First, plural sets of size information, each of which indicates longitudinal and lateral lengths of a sheet having a predetermined shape, are stored according to size. Next, a size of a document part of inputted image data is obtained. Then, the size of the document part is compared with the plural sets of size information. In a case where there is size information which differs from the size of the document part within an allowable range, the image data is cropped based on size information which indicates a size closest to the size of the document part. Meanwhile, in a case where there is no size information which differs from the size of the document part within an allowable range, the image data is cropped based on the size of the document part.